(2) State employment of women during the Ur III period 

Second and third REFEMA workshops, Tokyo (June 2013) and Carqueiranne (September 2013)

Bertrand Lafont (CNRS, Nanterre)

As a huge part of the Ur III documentation concerns the administrative management of state-dependent laborers and the calculation of what was allotted to them, we have of course much more information on women at work in the public sphere and outside their family than inside (cf. previous workshop).

1. MEN AND WOMEN AT WORK

We will start with a brief general “status questionis” on our subject, before having a look at some new evidence on results generated by the recently published archives of Garšana and Irisagrig.

In the written documents concerning the management of state-dependent laborers sustained through the so-called “ration system”, men and women were very often recorded together: thousands of texts record a joint administrative management of guruš and geme₂ side by side (in many administrative Ur III texts we can find for example accounts for n geme₂ guruš u₄ 1-šè, “n female and male work-days” counted together). It seems therefore that women contributed in the same way as men to the economic life managed by the state and we can emphasize right away that phenomenon of mixed management of human resources.

Nevertheless, in this regard, we must challenge the usual view according to which men worked for the primary productive sector of the economy (outdoors in the fields and in animal husbandry), while women were mainly occupied in the secondary productive sector (indoor processing activities to produce flour, oil, clothing, etc.)[1]. This gender dimension of labor division does not precisely fit our documentation, as we find in the following:

• male weavers (guruš-uš-bar) next to female weavers (geme₂-uš-bar), as in the tablet Legrain, UET 3, 1449
• male millers working alongside female millers (Watson, BCT 2, 49)
• or, on the contrary, many women involved in agricultural activities (passim).

In fact, it seems that we can divide the economic contribution of the female part of the working population between two kinds of activities:

• Outdoor activities: agriculture, hauling boats, making and carrying bricks, construction projects,  production and food processing. In these activities women contributed in the same way as men for the same kinds of collective work. They were employed in a large number of jobs, sometimes in unexpected sectors, as for example these female workers (geme₂) “having built the house of Amar-Suen” (é ^damar-^dsuen-ka dù-a, Sauren, NYPL 138). Perhaps these activities had something to do with the corvée-duty: according to P. Steinkeller[2], only heads of families owed such a service to the state but it is clear that women, needing more salaries and rations, were also involved in the same collective projects.
• Indoor activities: in this case, the professional specialization of women was most often indicated and three of these were more specifically their own domain: weaving (geme₂-uš-bar), milling flour (geme₂-kikken₂), and extracting oil (geme₂-ì-sur).

Concerning these indoor processing activities, as has been thouroughly studied (thanks to the works of Waetzoldt[3], Maekawa[4], Grégoire[5], Englund[6], and others), the real feature that was quite unique to this time is that some of them were clearly organized on an industrial scale. All around the country, some large factories employing female workers were operated especially for producing garments, flour and oil. And it is even possible that in or around the major cities of the kingdom, some individual “industrial” centers combining these various activities did exist:

• At Ur, female weavers, oil pressers and millers were allotted all together as can be seen in the tablet Legrain, UET 3, 1504, viii 28-29: níg-ba geme₂-uš-bar geme₂-ì-sur ù geme₂-kikken₂, “allotments for female weavers, oil pressers and millers”.
• At Garšana they built a unique triple complex, in the form of a “brewery-kitchen-mill building”: é-lunga é-kikken₂ ù é-muhaldim (index CUSAS 4, passim).
• And at the same place they built another double complex in the form of a “textile mill and craftsmen’s building”: é-uš-bar ù é-gašam-e-ne (index CUSAS 4, passim)[7].

Therefore, these main “industrial” factories employing women were weaving mills (é-uš-bar), places for pressing oil (é-ì-sur-sur), mills producing various flours (é‑kikken₂), kitchens (é‑muhaldim), and breweries (é-lunga). The following chart offers a summary.

The work of women in large textile factories (é-uš-bar) was intensively analyzed by Waetzoldt 40 years ago (UNT, p. 91-108) and that study remains the reference synthesis on this subject. As an illustration, let us consider a tablet kept in the British Musem (Maekawa, ASJ 20, 1998, 108 n°7). This enormous accounting text calculates the amount of barley distributed to personnel working in the various public weaving mills in the province of Lagash in the year Šulgi 48.

As analyzed by A. Uchitel[8], this text makes a clear distinction between 1) those receiving monthly rations because they were supposed to work only part-time in the workshop (note that some of them were receiving more than the usual average rations), and 2) those receiving yearly grain rations, likely as they did not belong to families holding allotment plots. In total, the text records more than 10,000 individuals working in the weaving mills of the province of Lagash in Šulgi 48. Almost ⅔ of them were female weavers but men, children and elders were also categorized, divided according to the quota of rations they received (from 100 liters to 10 liters per person and per month). Once again, these are amazing numbers.

And for an example of a mill producing flour and employing female workers, let us have a look at another text from the British Museum (King, CT 3, pl. 19-20, BM 18344). This is a comprehensive list of various workers receiving grain rations, and it can be completed based on other several similar large accounts from Girsu. The text provides a good example of how a millhouse was organized.

This “great millhouse” (é-kikken₂ gu-la) of Girsu seems to have employed 658 people in total. Two thirds of them were female millers, counted with their children. But some others were employed as weavers and oil pressers. Those in 1), 2) and 3) represent the workers in charge of producing flour. But the text also records people belonging to the regular staff in charge of maintenance and supervision of the mill, especially craft workers, boat towers, gardeners and specialists in boat constructions: most of them were male, recorded in 4), 5) and 6). We see therefore that, besides female millers who constituted the core of the workshop, a variety of craftsmen supported their activity.

But who were these state-dependents women working in outdoor activities or in these various mills? And what do we know about them or about their origin?

2. STATE DEPENDENT WOMEN

To explore this question, let us start with an important royal inscription of king Šu-Suen, after his victorious military campaign against Šimanum in the Zagros mountains (Frayne, RIME 3/2.1.4.3, 129’-145’). It is said there that:

« The king blinded the young men of those cities, whom he had overtaken. And the women of those cities, whom he had overtaken, he offered as a present to the weaving mills ».

As we can see, after this military campaign, male and female prisoners were separated. And whereas the male captives were blinded (probably to avoid any revolt without losing such a labor force), the female ones were not mutilated, but sent off to some textile mills of the inner kingdom. It is therefore clear that the labor force of Neo-Sumerian textile mills consisted in part of prisoners of war, a situation that is not without parallel, as for example at Mari in Old Babylonian time with its « ergasterions » (nepârâtum), that were large prison-workshops where mass productions were organized[9].

In brief, from all the documentation it appears that women working in these Ur III factories included:

• war-captives (nam-ra-ak)[10],
• purchased or indebted slaves (geme₂ // arad₂)
• donated personnel (a-ru-a)[11],
• women of the impoverished classes and outcasts of society.

Yet these categories of women certainly did not represent the entire workforce inside the state-run workshops. They would have been an insufficient number to meet the economic needs of the state. In all likelihood, many of the female workers employed in the mills were simply state-dependants, working part-time or full-time for rations and salaries, like their male counterparts.

M. Van de Mieroop[12] calculated that more than 13,000 female weavers were active at a single time in the workshops of Ur. Such a huge number is consistent with a text from Ur (Legrain, UET 3, 1504), where wool is distributed to female workers involved in the factories of that city for a total of 8,542 kg of wool. Waetzoldt calculated that at the rate of 1 mina of wool per capita (likely according to parallels), more than 17,000 female workers could have been concerned, employed as weavers, millers and oil pressers. In the same way Waetzoldt also calculated (UNT, p. 99) that, at Lagaš/Girsu, 15,000 personnel were employed in the weaving industries of the province.

Among them, 6,466 female weavers are explicitly recorded in two Girsu administrative texts dated to the year Amar-Suen 1 (Sigrist, SAT 1, 279 // Hussey, HSS 4, 3). These two tablets inform us that, among the three main urban centers of the province of Lagaš (Girsu, Kinunir and Gu-abba), the greatest weaving mills were situated in the third one, Gu-abba.

In total, thousands of women were therefore involved in every province of the kingdom and this must reflect a significant part of the population. However, it remains impossible to assess what percentage of the total female population was thus employed in these state-run workshops, or to estimate the importance of their contribution to the global economy, even if it may have been substantial. And it is also difficult to differentiate groups of women in terms of social hierarchy.

In several studies (and in a forthcoming article on “Corvée Labor in Ur III Times”)[13], P. Steinkeller has considered that two categories of workers must be distinguished in Ur III society: those belonging to the eren₂ class, who received royal land and rations in exchange for part-time service, and those who were semi-free workers (ùg-ÍL, geme₂), working all year round as unskilled laborers in exchange for rations. In all likelihood, men and women belonging to both of these categories were employed in the mills.

As for the unskilled workers, it is interesting to find (but not so frequently) in administrative texts some mentions of female menials (geme₂ ùg-ÍL) parallel to male ùg-ÍL. Sometimes their occupations are given: geme₂ ùg-ÍL mar-tu “menial Amorite women”, geme₂ ùg-ÍL amar-e gub-ba, “menial women on duty for the calves”, geme₂ ùg-ÍL é-gu₄-niga-ka “menial women of the house of fattened cattle”.

Another point can be observed: I. J. Gelb noticed[14] that, in many Ur III administrative texts, women are recorded alone with their children, without any accompanying man. He called these lists of rations for women and their children the “geme₂-dumu texts”, and he asked: “where are the husbands?”. We must confess that, 40 years later, this important question remains unanswered. We know almost nothing about the family life of these thousands of geme₂ employed in these large weaving and milling factories and working there with their children. Were there dormitories or barracks for them around the factories? Or did they return home every evening to their families and villages nearby? We have no information regarding these questions. But we can consider that, depending on the status of these women, individual situations probably varied: prisoners and slaves were probably confined to barracks, while “free” workers had their own home and family life independently, and came to work every day with their younger children. In any case, general conditions in these factories were harsh and restrictive, as we can see by the need of porters/keepers (ì-du₈) to control inputs and outputs, and the frequent mention of fugitives (zàh) that were pursued (the text Legrain, UET 3, 1018 records up to 134 of these geme₂-zàh!), or dead (úš) people on many occasions.

3. MANAGEMENT OF WOMEN AT WORK

For the management of full-year working women, the best and most frequently mentioned example is the one given by the tablet studied some twenty years ago by Robert Englund in his article “Hard Work” (see note 6). Instead of commenting for the umpteenth time on the content of this text, let us consider a parallel document from Umma: Genouillac, TCL 5, 5670. On the same model, it is a balance account written by a man named Ur-Šara, the foreman of a mill.

Over a period of one year, the state allocated to Ur-Šara various amounts of grain and a team of 36 female workers. This was his “debit” (meaning that “this was owed to the state”). The credit column describes all the work performed by these 36 female workers during the year: two thirds of their time was spent producing flour. Yet it is interesting to observe that the remainder of their time was occupied in agricultural work, in weaving, in loading and unloading boats, in producing earthworks, and so on. One sixth of the time was counted as days off (that is 5 days per month, 2 months per year). Finally, all of this was converted in the final balance into equivalents of workdays and of produced flour. Compared to what was expected, there was a deficit, to be reimbursed by these women in the following year.

Here, we can observe that these women were employed full-time, all year long in the mill. But it is noteworthy that they could easily be assigned to occupations besides producing flour, to suit the needs of the economic organization.

We just saw above that, in total, these female workers were counted in the thousands for each province. Consequently, with the Ur III accounting system and its “workdays” equivalencies, some summary administrative texts finally calculated women workforce in tens of thousands of female “workdays” (geme₂ u₄ 1-šè). Below is a chart that shows the greatest numbers of workdays of female workers that can be found in Ur III texts. And one will not be surprised to see, once again, that milling and textile production are principally concerned.

How were these women paid? As it is well known, the “ration system” with its amounts of barley, oil and wool issued to dependents, was a hallmark of early Mesopotamian history and is especially well documented in the Ur III period: it is mainly because of such rations that we have such a huge written documentation. In this system, women were paid distinctly less than men, a phenomenon already attested in earlier periods (Ebla, Lagaš II): while the guruš received an average of 60 liters of grain per month, women usually received only 30 to 50 liters.

But as we just saw, only some people worked full time for the state and we can imagine that many of these women were married or related to members of the eren₂ class: they belonged to families that held plots of subsistence land and they could have stayed at home or worked only part-time in public factories. And as a matter of fact, we do find texts in which women could, like men, be “hired” (hun-gá) for some periods of time (an example of a text with geme₂ hun-gá, “hired women”, is Ozaki-Sigrist, BPOA 1, 564), and many texts record real “salaries for women” (á geme₂) instead of rations.

A last observation will be made about positions of authority that these women could hold in such a context of “public” work. It is not infrequent to find forewomen at the head of teams of workers, as for example in the text Waetzoldt-Yıldız, MVN 16, 727, 11-12. Sometimes women could hold high administrative positions, at the same level as men and could seal tablets, as shown by several seal impressions and the attested title of munus-dub-sar (as in Forde, Nebraska 45). And this situation can now be best observed in the archive of Garšana, where the management of the estate which was located there was greatly between the hands of women.

4. NEW DATA FROM GARŠANA AND IRISAGRIG

This leads us finally to the two large new archives that have recently been recovered, gathered, studied and published by David Owen: those of Garšana (published 6 years ago)[15] and those of Irisagrig (just published in Nisaba 15)[16]. In spite of their lack of archaeological context, these archival texts give unique insights into the culture of Sumer in the late 3^rd millenium. And they provide, for our REFEMA topic, unprecedented data on the role of women in the Neo-Sumerian economy and society.

Let us illustrate here this new wealth of information with a few brief examples. First, from the Garšana archive: this archive not only offers 1500 additional common Ur III administrative texts but it also provides the unique records of a huge and hitherto unknown private household: during the reigns of Šu-Suen and Ibbi-Suen, this household was the estate of a general (šagina) and physician (a-zu), named Šu-Kabta, and his wife, a princess named Simat Ištaran who was a sister of king Šu-Suen. This archive highlights espacially the role of women, both as heads of the household and in the capacity of supervisors, administrators (some of them have seals), and laborers in many sectors of the estate. And if one takes the point of view of the labor force available or used in this household, it is interesting to see that, in the all archive, working women topped in numbers working men. Finally, thanks to the new evidence of Garšana, we can see that women in Ur III society had more authority and influence than may have been previously recognized and sheds light on a hitherto undocumented aspect of society in the Ur III period.

After Šu-Kabta’s death, his wife, princess Simat-Ištaran, inherited and actively managed the Garšana estate herself. From that time on, several women of the household began to play an important role. As an illustration, let us consider the case of a woman named Aštaqqar who ended up managing the all textile industry at Garšana as supervisor of the weavers (ugula-uš-bar, a function usually occupied by men). She had her own seals, as shown below, with which she sealed tens of tablets (seals drawn by R. Mayr, CUSAS 3, p. 431).

Various other women also supervised construction gangs consisting of both men and women : such positions were until now unattested within the existing Ur III records[17].

As observed in my previous paper for the first REFEMA workshop at Nanterre, a quick look at the occupations attested in the household headed by Šu-Kabta and Simat-Ištaran presents at first the usual list for women : oil pressers, grain millers, or weavers. But it is noteworthy that these women were in fact often performing tasks far from their primary specialty. Consider for example the women who are qualified as weavers (geme₂-uš-bar). Even if they were surely mainly employed in the textile mill (é-uš-bar) of Garšana, we can also see them in many other occupations; they are thus employed for :

– Harvesting (CUSAS 3, 262 : 19-22) :

• 18 geme₂ sag-dub 1 geme₂ á-⅔ geme₂-uš-bar-me-éš (…) u₄ 105-šè (…) še gur₁₀-gur₁₀-dè zàr tab-ba-dè ù geš ra-ra-dè a-šà Ṣe-ru-um gub-ba
  « 18 full-time female weavers (and) 1 female weaver at ⅔ wages for 105 days (…) employed at the Ṣerum field to harvest barley, stack (its) sheaves (for drying) and flail (them) »

– Demolishing a house (CUSAS 3, 273 : 1-6) :

• 6 guruš ašgab 8 guruš lú-azlag₂ 24 geme₂-uš-bar u₄ 1-šè al-tar é A-bí-a-ti gul-dè gub-ba
  « 6 leather workers, 8 fullers (and) 24 female weavers employed for 1 day of construction to demolish the house of Abiati »

– Transporting crops of barley (CUSAS 3, 290 : 1-4)

• 16 ⅓ geme₂-uš-bar u₄ 10-šè še Ù-dag-ga^ki-ta GAR-ša-an-na^ki-šè de₆-a
  « 16 ⅓ female weavers for 10 days having brought barley from Udaga to Garšana »

– Lamenting after the death of their master Šu-Kabta (and they are paid for that, CUSAS 3, 252 : 1-7 et 257 : 1-6)

• 97 geme₂ sag-dub 15 geme₂ á-⅓-ta geme₂-uš-bar-me-éš [n] geme₂ [u₄] 9 ½-šè [gaba] sìg-dè gub-[ba u₄ Šu]-kab-tá ba-úš-a
  « 97 full-time female weavers (and) 15 female weavers at ⅓ wages (and) [n] female spinner(s) employed for 9 ½ [days] to beat (their) [breasts when] Šu-Kabta died »

As we can see, these specialized women were finally employed for many tasks far beyonf their basic specialty and mostly at the same level as men.

But at Garšana, one point is particularly striking: the involvement of women in the construction works sector. This has been recently and extensively studied by W. Heimpel[18] who has gathered especially more than 200 tablets that are the daily records of an ambitious construction project during the three last years of the life of Šu-Kabta, at the end of the reign of Šu-Suen. It seems that this construction has mainly concerned the building of a “ringwall”, of several factories and of private residences. In this context, it appears that women of the household were primarily involved in these construction work and did, for example, almost all the carrying of bricks (such a task was hitherto unattested elsewhere). Note the scheme in the last lines of many of these texts that list female brick carriers by name in groups headed by a foreman or a forewoman (table borrowed from W. Heimpel [19]).

These texts, sometimes having more than 100 lines, give at the end the number of carried bricks, the distance over which they were carried, the work standard in the form of numbers of carried bricks that earned a day’s wage, the amount of the wage, the total of barley required for pay, and the name of the overseer. As an illustration, consider one of these tablets : Owen, CUSAS 3, 352.

Rev.
7.	šu+nigin2 10 ⅓ sar 2 ½ gín sig4	Total 10,…  plots of bricks
8.	ús-bi 50 ninda	The (transportation) distance is 50 rods
9.	á geme2 1-a 216 sig4-ta	Work for one female worker: 216 bricks each (day)
10.	geme2-bi 34 ½ 5 gín	The number of maids (for one day) is 34,5…
11.	še 3 sila3-ta ba-hun	They are hired at 3 liters of barley each
12.	še-bi 0,1.4. 3 ⅔ sila3 5 gín	The needed barley is 103,…  liters
13.	á sig4 ÍL-gá é-šu-x-[…]	Wages for carrying bricks at GN.
14.	Me-é-a ugula lú hun-gá	ME-ea supervisor of hired persons
15.	giri3-bi [dIškur-illat] šabra	Responsible : Adad-tillati the administrator
16.	ù [Puzur4–dnin-kar-kè dub-sar]	and Puzur-Ninkarak the scribe.
17.	šà [GAR-ša-an-naki]	At Garšana
18.	iti […]	Date
19.	(Year Šu-Sîn 8)

Only women were concerned in these texts. Yet finally, what can be learned from the Garšana archive as a whole is the involvement of the women of the estate in all kinds of works: they were employed in the same way as men and even more than them, and these texts show how great was the flexibility in the organization of their work, regulated by bureaucratic authorities. This observation perhaps addresses the question of the supposed scarcity of labor in the Ur III period (an idea put forward by Piotr Steinkeller[20]). We can even ask about the existence of a kind of “labor market”  because of so many people mentioned in this archive as having been “hired” (hun-gá).

But a new archive adds now to this picture: the Irisagrig archive recently published by David Owen in two new volumes of the series Nisaba (volume 15/1 and 15/2)[21]. The 1,200 tablets published and studied there represent another example of a unique archive, this time from a major city of the Ur III kingdom: Irisagrig, not far from Nippur[22], known to have been the seat of a governorship. These texts are part of a large, official archive (probably of the provincial governor, named Ur-mes), and concern the affairs of various officials, mainly during the reign of Šu-Suen.

The identification and recovery of this large body of new data add a significant dimension to our understanding of the historic, religious, political, and socio-economic dynamics of the Ur III state. Like the Garšana texts, they add especially to the picture we can sketch for the role played by women in Ur III society. The number of named women in this archive is substantial and includes many newly attested (mostly Akkadian) female personal names, newly documented princesses, and wives of prominent officials, along with women in positions that are rarely, if ever, attested in other sources.

A more comprehensive study of the data will surely come in the near future. But we can have a quick look at these occupations, which include:

– “Usual” female occupations in the documents of Irisagrig:

• geme₂-àr-ra / geme₂-kikken₂, “millers”
• geme₂-uš-bar, “weavers”
• geme₂-kar-kíd, “single women, prostitutes” (?)

– “Unusual” female occupations in the documents of Irisagrig (references to the texts in the indices of Nisaba 15/1):

• geme₂-azlag₂ (usually male lú-azlag₂, “fuller, washerman”)
• munus-a-zu (usually male a-zu, “physician”)
• munus-dub-sar  (usually male dub-sar, “scribe”)
• munus-gudu₄ (usually male gudu₄, “purification priest”)
• geme₂-ì-du₈, munus-ì-du₈ (usually male ì-du₈, “doorkeeper”)
• geme₂-muhaldim, munus-muhaldim (usually male muhaldim, “cooker”)
• geme₂-nar, munus-nar (usually male nar, “singer, musician”)
• qadištum (written ^munusqá-dì-iš-tum)  nu-gig/qadištu, “midwife?”, was only attested until now in the recently published new parts of the Ur-Namma Code[23]

Such an amazing list surely deserves further study. But in the frame of this short paper, we will just examine, to conclude, an emerging Irisagrig batch of tablets that provides new data on the role of a new prominent woman at the head of a large household: this woman was named Ninsaga and was the daughter of the governor (ensi₂) of Irisagrig, Ur-mes. She was at the head of a large estate in that city. Three texts deserve attention: the first (Owen, Nisaba 15, 953), is a complete inventory, on twelve columns, of hundreds of named male and female slaves (arad₂ and geme₂) belonging to the household of Ninsaga in year Ibbi-Sîn 3. Each of them received monthly between 10 and 75 liters of barley rations. At the end, we find recorded: 74 guruš, 48 geme₂, 47 dumu-nita₂, 56 dumu-munus (122 adults, 103 children, that is 225 men and women). All these dependents of this private household are clearly characterized as “slaves” (geme₂-arad₂) of Ninsaga.

One year before (Ibbi-Suen 2) another text (Owen, Nisaba 15, 797) shows that a general (šagina) had taken away 50 slaves who had been removed from this household of Ninsaga, probably to bring them to the king, since these slaves are characterized as being assigned to the throne (arad₂-geme₂ nam-^gešgu-za). And it is interesting to find again this same information exactly quoted in another text (Owen, Nisaba 15, 1038), which summarized other withdrawals made over Ninsaga’s household staff.

Finally, it appears that, if 51 slaves were left (si-ì-tum) in the household of Ninsaga after the “removal” carried out by the military officer, the large census that we mentioned at the beginning shows that, one year after (in Ibbi-Suen 3), the staff of the household had increased to 122 slaves.

What we see in the end and for the first time with this small batch of tablets, is the great number of dependents, men and women, that could be found in a private household and how these dependents could be added or removed from the household to be posted elsewhere, according to the will of administrative authorities. And this could maybe makes us reconsider what could have been the exact status of these slaves and dependents in these “private” households during the Ur III period.

PROVISIONAL CONCLUSIONS

Some provisional main points can be summarized:

During the Ur III period, women were not confined to the spheres of family, marriage and household: they were employed along with men in all kinds of indoor or outdoor work and activities.

Milling, weaving and pressing oil were essentially state-run activities, organized on an “industrial” scale and performed by state-dependent women working in large production units to meet the huge needs of the State for its “ration allotment” system. Weavers, millers and oil pressers were often associated. Their installations, perhaps aggregated in unique places, seem to have been managed under the direct supervision of the royal power.

We should beware of ​​a depreciating vision according to which only poverty or enslavement could explain and justify the work of women. Texts show the same intensive use of male and female workers belonging to various categories of the society in the all kingdom.

The pressure put by the state administration and its strong will of control observable through the texts should probably be analyzed in the context of a scarcity of labor during the Ur III period, an idea previously suggested by P. Steinkeller.

Finally, one major gap characterizes our documentation: there is absolutely no information in any of our texts regarding the potential participation of women in market activities. Is it simply because, in this area, written documents were usually unnecessary or outside the administrative sphere?

---

---

---